Pith. sign in

REVIEW 4 major objections 6 minor 11 references

Performance analysis of IEEE 802.11ax heterogeneous network in the presence of hidden terminals

T0 review · 4 major / 6 minor · reviewed 2026-08-14 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read The paper argues that raising the carrier sensing threshold of stations during association with an IEEE 802.11ax access point can reduce uplink collisions from hidden terminals enough to restore the throughput of a no-hidden-terminal…

desk verdict Incremental expansion of the authors' conference paper; the main simulation claim is confounded because CSTH and node density vary together, so the throughput recovery cannot be attributed to the proposed mechanism. read the letter →

arxiv 1908.01834 v1 pith:BQ7SJHQD submitted 2019-08-05 cs.NI

M. Zulfiker Ali , Jelena Mišić , Vojislav B. Mišić This is my paper · ORCID
classification cs.NI
keywords hiddenterminalscarriersensingthresholdIEEE802.11axuplinktransmissionMU-MIMOEDCAnetworkthroughputcollisionprobability
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

Hidden terminals are stations that cannot hear each other but can collide at the access point; in IEEE 802.11ax uplink traffic they cause retransmissions, waiting delays, and throughput loss that trigger-based multiuser transmission cannot fully remove. The paper argues that a simple association-time adjustment—raising each station's carrier sensing threshold from -82 dBm to -73 dBm—shrinks the AP's effective range so that every associated station lies within hearing range of every other. In the paper's simulated 24-station network at 4800 packets/sec, this change cut the average number of hidden nodes per station from about 4 to 0.25 and raised uplink throughput from 357 Mbps to 438 Mbps, matching the no-hidden-terminal case. The authors also propose specification changes to keep adjacent basic service sets on different primary channels and to restrict NAV updates to same-BSS transmissions, preventing inter-BSS interference that the asymmetric sensing ranges could create.

What carries the argument

The carrier sensing threshold (CSTH), the minimum received signal power a station treats as a busy channel, is the mechanism. During association the paper raises CSTH from -82 dBm to -73 dBm, which reduces the AP's transmission radius and therefore the set of stations that can associate; with enough station density, the associated set becomes one in which every pair can hear each other. The analytic model expresses the success probability of each access category as the no-hidden-node success probability multiplied by a factor $f_{ncoll}=f_h^{(1-f_{mu})T_{v-su}+f_{mu}T_{v-mu}}$ that accounts for hidden-node transmissions during the vulnerable period, where $T_{v-su}=rts+sifs+cts$ and $T_{v-mu}=trig$.

What would settle it

Repeat the simulation at 4800 packets/sec with a fixed station density and record the number of associated stations and uplink throughput as CSTH rises from -82 dBm to -73 dBm; if throughput does not reach the no-hidden-terminal value of about 438 Mbps, or if fewer than 24 stations can associate, the claim fails.

Watch

Extended reading notes

Core claim

The central discovery is that the hidden-terminal degradation of uplink IEEE 802.11ax throughput is not an unavoidable property of coexistence between HE and legacy devices; it can be largely removed by making the association process choose a smaller, mutually audible set of stations. The load-bearing mechanism is the carrier sensing threshold: by raising CSTH during association only, the AP's effective transmission range shrinks, so the stations that remain associated are physically closer to the AP and to each other. During normal operation the threshold stays at the legacy value, so the network does not sacrifice spatial reuse after association. Simulation shows that collision count, packet drop (about 16% at 4800 packets/sec with hidden nodes), waiting time, and backoff time all fall toward the no-hidden-terminal baseline as CSTH rises, with -73 dBm restoring throughput to 438 Mbps.

Load-bearing premise

The simulation assumes station density rises to keep exactly 24 stations associated as the AP's range shrinks; in a fixed-density network, raising the threshold would associate fewer stations and the claimed throughput recovery might not follow.

Editorial extensions

If this is right

  • At 4800 packets/sec, raising association-time CSTH from -82 dBm to -73 dBm raises uplink throughput from 357 Mbps to 438 Mbps, matching the no-hidden-terminal network.
  • The average number of hidden nodes per station drops from about 4 to 0.25 over the same threshold change.
  • Packet drop at high load, around 16% with hidden terminals, falls to almost zero once CSTH reaches -73 dBm.
  • Keeping 24 associated stations requires raising station concentration as the AP's range shrinks, so the benefit is demonstrated for dense deployments.
  • Adjacent BSSs operating on different primary channels and same-BSS-only NAV updates would be needed to prevent inter-BSS interference from the asymmetric transmission ranges.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • A direct deployment rule would be to raise the association-time threshold only until a measured hidden-node collision rate falls below a target, since the paper's fixed 10 dB increase is tied to a specific density and range.
  • In a fixed-density deployment the same 10 dB increase would trade coverage for throughput, so the stated recovery should be read as a density-dependent result rather than a universal recommendation.
  • The proposed NAV restriction could be tested separately from the CSTH change: a simulation with both mechanisms turned off and on independently would show how much of the improvement comes from each.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 6 minor

Summary. The paper studies the impact of hidden terminals on uplink transmissions in an IEEE 802.11ax heterogeneous network with both HE and legacy stations. It proposes increasing the carrier sensing threshold (CSTH) of STAs during association with an HE access point, arguing that this shrinks the effective BSS footprint so that associated STAs are more likely to hear each other, thereby reducing hidden-terminal collisions. The paper gives an analytical model in Section 3, with equations for idle-slot probabilities, successful transmission probabilities, and hidden-node collision probabilities, and then presents event-driven simulation results in Section 4. The central numerical claim is that at 4800 packets/s, raising the association CSTH from -82 dBm to -73 dBm improves throughput from 357 to 438 Mbps, matching the no-hidden-terminal scenario, and reduces the average number of hidden nodes per STA from about 4 to 0.25. The paper also proposes three protocol modifications: higher CSTH during association, adjacent BSSs on different primary channels, and NAV updates only from same-BSS transmissions.

Significance. If the claimed effect is real, the proposal is practically valuable because it is a simple, standard-compatible change to the association procedure rather than a new access protocol. The simulation study covers a useful range of metrics (backoff time, waiting time, collisions, packet drop, throughput, overhead, hidden-node count) and includes a direct comparison with and without hidden terminals, which strengthens the descriptive part of the work. The main claims, however, rest on the CSTH sweep in Section 4, and that sweep is confounded by an accompanying increase in station density, as the authors themselves state. The analytical model in Section 3 is imported from prior work and contains a mathematical error in Eq. (5), and it is never validated against the simulation. These issues make the central quantitative claims currently unsupported.

major comments (4)
  1. [Section 3.1, Eq. (5)] Equation (5) is mathematically incorrect as written. The probability that no hidden node transmits in a given slot should be a product over traffic classes of (1 - tau_k)^{N_{k,h}}, not a sum over classes of (1 - tau_k)^{N_{k,h}} tau_k^{N_{k,h}}. The extra factor tau_k^{N_{k,h}} has no justification, and the summation over classes is inconsistent with the requirement that every hidden node in every class refrain from transmitting. Because f_h is then used in Eq. (6) to compute f_ncoll and hence gamma_{k,h}, this error propagates through the entire analytical model. The authors should correct Eq. (5) and, ideally, compare the corrected f_ncoll against the simulated collision counts in Figs. 13 and 14.
  2. [Section 4, simulation setup and Figs. 9, 12, 14, 16, 19, 22] The central CSTH sweep changes two variables at once. The text states: 'when we increase the carrier sensing threshold of STAs during association process, the transmission range of AP is actually reduced. Therefore, we increase the concentration of STA to ensure that we have same number of STAs associated with AP.' Thus, as CSTH increases, the AP coverage shrinks and the same 24 STAs are packed into a smaller area, shortening link distances and increasing connectivity. The reported improvements in backoff, collisions, packet drop, throughput, and hidden-node count could therefore be driven largely by the density compensation rather than by the association-CSTH mechanism itself. In a fixed deployment, raising the association CSTH would instead exclude distant STAs from the BSS, and the paper does not model that coverage/offload tradeoff. The claimed throughput recovery from 357 to 438 Mbps is consequently not isolated to the proposed mechanism. The authors should either run a fixed-density scenario and report how many STAs associate at each CSTH, or add a control scenario in which CSTH is varied without density compensation.
  3. [Section 3.2] The claim that increasing the association CSTH 'ensure[s] that all STAs within the BSS are also within the transmission range of each other' is geometrically under-justified. Two STAs near opposite edges of the reduced AP transmission range can be separated by up to roughly twice the AP range, which can exceed the STA-to-STA transmission range even when both are associated with the AP. The authors' own Fig. 22 shows a residual average of 0.25 hidden nodes per STA at -73 dBm, which is inconsistent with the claimed complete elimination. The statement should be weakened to a statistical reduction, or the geometry should be analyzed with an explicit condition on STA ranges relative to the BSS diameter.
  4. [Section 3 and Section 4] The analytical model of Section 3 is not used or validated in the simulation section. Equations (1)-(7) define gamma_k, f_h, f_ncoll, and gamma_{k,h}, but Section 4 never compares these predictions with the simulated collision probabilities or throughput values. Given that Eq. (5) contains the error noted above, the analytical contribution is currently disconnected from the numerical results and cannot be checked. The authors should either remove the analytical model from the claims or provide a direct comparison between the corrected model and the simulation output.
minor comments (6)
  1. [Section 3.2] There is a typo: 'A prpmising technique' should be 'A promising technique'.
  2. [Section 3, Fig. 4 caption and text] The text below Fig. 4 says 'the transmission range of the APP and all other nodes'; 'APP' should be 'AP'.
  3. [Figs. 17-19] The y-axis labels in Figs. 17-19 read 'MB', while the text reports throughput in Mbps. The units should be made consistent and clearly defined.
  4. [Fig. 22] Fig. 22 shows an average number of hidden nodes per STA, but no confidence intervals or per-run variability are reported even though the simulation was averaged over 10 runs. A brief statement of the spread would help assess whether the decrease to 0.25 is statistically meaningful.
  5. [Section 4, first paragraph] The station placement procedure is not fully specified. The text says STAs are placed 'randomly on the same floor', but does not state the area size, the minimum inter-STA distance, or how the concentration is adjusted when CSTH changes, which is important for reproducibility of the density-compensated sweeps.
  6. [Section 2] The term 'heterogeneous network' is used throughout, but the only heterogeneity described is the coexistence of HE and legacy STAs in a single BSS. The authors should clarify whether they intend a multi-BSS deployment, since the proposed inter-BSS mitigations in Section 3.2 are never simulated.

Circularity Check

1 steps flagged · score 6.0 of 10

Throughput recovery at CSTH -73 dBm is confounded with the density compensation imposed by the simulation setup, so the hidden-node reduction is an artifact of construction.

  1. fitted input called prediction [Section 4, simulation setup paragraph (page 7); results reported in Figs. 19 and 22]
    "The concentration of the nodes is controlled so that the AP has always 24 nodes associated with it. For each simulation setup, when we increase the carrier sensing threshold of STAs during association process, the transmission range of AP is actually reduced. Therefore, we increase the concentration of STA to ensure that we have same number of STAs associated with AP."

    The paper varies CSTH by shrinking the AP radius and, at the same time, increases STA concentration to hold the association count at 24. Hidden-node incidence depends on inter-STA spacing for a fixed carrier-sense range; repacking the same 24 STAs into a smaller disk by construction reduces hidden pairs. The paper then reports this geometrically forced reduction as confirmation of the association-CSTH mechanism (hidden nodes per STA from about 4 to 0.25; throughput from 357 to 438 Mbps at 4800 packets/s). Because the density change is an input imposed by the authors, the 'prediction' is not an independent test of the mechanism; a fixed-deployment experiment with fewer associated STAs would be needed to separate coverage offload from the claimed benefit.

full rationale

The only load-bearing circular step is in the simulation design. Section 4 states that when CSTH is raised during association, AP range is reduced, and the authors increase STA concentration to keep 24 associated STAs. Thus the CSTH sweep varies spatial density together with the threshold. Hidden-node count is a geometric function of inter-STA distances for a fixed carrier-sensing range, so packing 24 STAs into a shrinking area by construction lowers the hidden-node count and restores throughput; the headline numbers are largely the result of this imposed geometry. This is not malicious fitting, but it is a case where the 'prediction' is built into the input scenario, hence partial circularity. The analytic model in Sec. 3.1 uses fmu from the authors' prior work [6], but the simulation results do not depend on that model, and the CSTH recommendation is read off the simulation sweep, so the self-citation is not load-bearing. No uniqueness theorem or ansatz is smuggled in. Had the paper simulated a fixed deployment and varied only association CSTH, accepting fewer associated STAs or explicitly modeling offload, the comparison would be an independent test; as written, the claimed benefit reduces to the construction.

Assumptions & free parameters 2 free parameters · 4 assumptions · 0 invented entities

The analytical portion leans on the authors' prior EDCA and uplink MU models, and the hidden-terminal reduction is partly built into the simulation by shrinking the AP range while repopulating the cell with denser STAs. The central performance numbers come from a single custom simulator with no released artifacts or independent validation.

free parameters (2)
  • CSTH increase during association = -73 dBm (10 dB above legacy -82 dBm)
    Chosen from the simulation sweep in Section 4; the recommendation is based on the observed threshold at which backoff time, collisions, and throughput match the no-hidden-terminal case.
  • STA density adjustment = not quantified
    When CSTH is raised, the simulation increases STA concentration to keep 24 associated nodes; this is an ad hoc setup that confounds the CSTH effect with spatial density.
assumptions (4)
  • domain assumption All nodes are within the AP's transmission range, so there are no hidden terminals from the AP's perspective.
    Section 3: 'Since all the nodes are associated with the AP, we can safely assume that all the nodes are within the transmission range of the AP.' This underpins the claim that only the trigger frame and RTS are vulnerable to collision.
  • domain assumption RTS/CTS and NAV updates protect data frames in the uplink from hidden-node collisions.
    Section 3 and the discussion of Fig. 23: 'as soon as AP transmits CTS, the NAV of all other STAs are updated which eliminates hidden node problem for data packets.' The simulation relies on this to keep collisions limited to RTS/CTS and trigger frames.
  • ad hoc to paper The analytic hidden-node model of Eqs (4) to (7) from the authors' prior work [6] applies to the simulated scenario.
    Eqs (4) to (7) are presented without derivation, with fmu sourced to [6]. Eq (5) as written contains a summation/product error, so the analytic support is not self-contained or internally consistent.
  • ad hoc to paper The proposed inter-BSS mitigations (different primary channels and NAV filtering) can be realized without breaking 802.11ax compatibility.
    Section 3.2 lists these as modifications to the draft specification, but they are not simulated or analyzed. The paper admits the approach 'may also lead to inter-BSS interference issue' and then asserts it can be easily eliminated.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Performance analysis of IEEE 802.11ax heterogeneous network in the presence of hidden terminals." pith.science (2026). https://pith.science/paper/BQ7SJHQD

@misc{pith2026190801834,
  author       = {Pith},
  title        = {Pith review of: Performance analysis of IEEE 802.11ax heterogeneous network in the presence of hidden terminals},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/BQ7SJHQD}},
  note         = {Machine review of arXiv:1908.01834}
}
read the original abstract

Performance improvement has been among the foci of all previous amendments of IEEE 802.11 protocol. In addition, the draft high efficiency (HE) amendment IEEE 802.11ax, proposed by TGax, aims at increasing network performance. One of the main obstacles to improving spectral and power efficiency is the presence of hidden terminals which degrade throughput, in particular in uplink transmission. IEEE 802.11ax does provide mechanisms such as trigger based uplink transmission that mitigate this degradation to some extent, but are incapable of eliminating it, esp. at high arrival rates. To combat the hidden terminal problem, we propose to increase the carrier sensing threshold (CSTH) of STAs during association with an HE access point. Our results confirm that the proposed mechanism can lead to significant reduction of collision probability in uplink transmission.

Figures

Figures reproduced from arXiv: 1908.01834 by the authors.

Figure 1
Figure 1. EDCA channel prioritized access (adopted [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 3
Figure 3. Basic Service Subset. AP Node B Node A Node C Tx range of Node A Tx range of AP Tx range of Node C [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figure 7
Figure 7. Although multiple uplink transmissions take place, due to collision from hidden terminals we observe a [PITH_FULL_IMAGE:figures/full_fig_p007_7.png] view at source ↗
Figures from the paper (8 more)
Figure 5
Figure 5. Figure 5: Path Loss (proposed 11ax standard). 0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 dB Distance(m) CSTH [PITH_FULL_IMAGE:figures/full_fig_p008_5.png]
Figure 7
Figure 7. Figure 7: Backoff time, hidden node present. 0 200 400 600 800 1000 1200 1400 1600 0 1000 2000 3000 4000 5000 6000 7000 8000 Musec Offered load (packet) Backoff time AC0 AC1 AC2 AC3 [PITH_FULL_IMAGE:figures/full_fig_p008_7.png]
Figure 12
Figure 12. Figure 12: Waiting time in the queue with CSTH increased to 4800 pack￾ets/sec. packets/sec), many STAs need to wait for the medium for a longer time and consequently the backoff time increases, as shown in [PITH_FULL_IMAGE:figures/full_fig_p008_12.png]
Figure 14
Figure 14. Figure 14: Number of collisions with CSTH increased [PITH_FULL_IMAGE:figures/full_fig_p009_14.png]
Figure 16
Figure 16. Figure 16: Packet drop with CSTH increased to 4800 packets/sec. terminal is much higher than the waiting time of a packet without hidden terminals. The decrease of waiting time with the increase of CSTH is shown in [PITH_FULL_IMAGE:figures/full_fig_p009_16.png]
Figure 17
Figure 17. Figure 17: Network throughput, hid￾den node present. 0.00 100.00 200.00 300.00 400.00 500.00 600.00 700.00 0 1000 2000 3000 4000 5000 6000 7000 8000 MB Offered load (packet) Throughput [PITH_FULL_IMAGE:figures/full_fig_p010_17.png]
Figure 20
Figure 20. Figure 20: MAC overhead, hidden node present. 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0 1000 2000 3000 4000 5000 6000 7000 8000 % Offered load (packet) Overhead [PITH_FULL_IMAGE:figures/full_fig_p010_20.png]
Figure 22
Figure 22. Figure 22: Average number of hidden nodes per STA. 5400 5500 5600 5700 5800 5900 6000 6100 6200 6300 6400 0 2000 4000 6000 8000 10000 12000 14000 No. of collisions Packet Length (byte) [PITH_FULL_IMAGE:figures/full_fig_p011_22.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

11 extracted references · 11 canonical work pages

  1. [1]

    M. Z. Ali, J. Mi ˇsi´c, and V . B. Mi ˇsi´c. Impact of hidden nodes on uplink transmission in IEEE 802.11ax heterogeneous network. In Int. Wireless Communications and Mobile Computing Conf. (IWCMC) , Limassol, Cyprus, June 2018

    work page 2018

  2. [2]

    Al-Bado, C

    M. Al-Bado, C. Sengul, C. J. Sreenan, and K. N. Brown. Hidden terminal management for uplink traffic in rate-controlled WiFi networks. In 2016 IEEE Symposium on Computers and Communication (ISCC) , pages 1066–1071, June 2016. 11

    work page 2016

  3. [3]

    Khorov, A

    E. Khorov, A. Kiryanov, and A. Lyakhov. Ieee 802.11ax: How to build high efficiency wlans. In 2015 International Conference on Engineering and Telecommunication (EnT) , pages 14–19, Nov 2015

    work page 2015

  4. [4]

    A. k. Ajami and H. Artail. On the modeling and analysis of uplink and downlink ieee 802.11ax wi-fi with lte in unlicensed spectrum. IEEE Transactions on Wireless Communications, 16(9):5779–5795, Sept 2017

    work page 2017

  5. [5]

    M. Z. Ali, J. Mi ˇsi´c, and V . B. Miˇsi´c. Performance analysis of downlink MU-TXOP sharing in IEEE 802.11ac. IEEE Transactions on V ehicular Technology, PP(99):1–1, 2017

    work page 2017

  6. [6]

    M. Z. Ali, J. Mi ˇsi´c, and V . B. Miˇsi´c. Uplink Access Protocol in IEEE 802.11ac. IEEE Transactions on Wireless Communications, 17(8):5535–5551, Aug 2018

    work page 2018

  7. [7]

    S. I. Sou and Y . Lee. Trigger-based approach with hidden node problem for uplink multi-user transmission in 802.11ax. In 2017 IEEE 18th International Workshop on Signal Processing Advances in Wireless Communi- cations (SPA WC), pages 1–5, July 2017

    work page 2017

  8. [8]

    Indication for UL MU carrier sensing

    Kiseon Ryu. Indication for UL MU carrier sensing. URL:https://mentor.ieee.org/802.11/dcn/16/11-16-0057- 00-00ax-indication-for-ul-mu-carrier-sensing.pptx. 2016

    work page 2016

Show all 11 references
  1. [9]

    Misic, S

    J. Misic, S. Rashwand, and V . B. Misic. Analysis of Impact of TXOP Allocation on IEEE 802.11e EDCA under Variable Network Load. IEEE Transactions on Parallel and Distributed Systems , 23(5):785–799, May 2012

    2012

  2. [10]

    IEEE standard 802.11ax, The IEEE 802.11 Working Group of the 802 Committee, USA, March 2016

    Wireless LANs: Proposed TGax draft specification. IEEE standard 802.11ax, The IEEE 802.11 Working Group of the 802 Committee, USA, March 2016

    2016

  3. [11]

    Mvulla, E

    J. Mvulla, E. C. Park, M. Adnan, and J. H. Son. Analysis of asymmetric hidden node problem in IEEE 802.11ax heterogeneous WLANs. In 2015 International Conference on Information and Communication Technology Convergence (ICTC), pages 539–544, Oct 2015. 12

    2015

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.