CBAND: Multifaceted Spectrum and Network Optimization

• CBAND is a multi-domain term describing C-band optical networks, C-BASS radio surveys, cellular spectrum allocations, and channel bonding in WLANs.
• In optical networks, CBAND underpins congestion management with techniques like the DACA algorithm reducing blocking probabilities by up to 74% and enhancing channel utilization.
• Across fields, CBAND supports 5 GHz cosmic surveys, cellular performance optimizations, and efficient spectrum sharing, highlighting its technical versatility.

In the cited literature, CBAND is not a single universally fixed term but a domain-dependent label spanning several technically distinct meanings. It denotes the C band in optical transport networks, the C-Band All-Sky Survey (C-BASS) in radio astronomy and CMB foreground analysis, the conventional cellular band in cognitive cellular systems, and channel bonding (CB) in WLAN and cognitive-radio sensor-network contexts. It also appears in studies of C-band 5G/CBRS, C-band satellite–terrestrial coexistence, and extended C-band VSAT links. The commonality is spectral-resource management in a frequency band or bonded channel set, but the underlying physical systems, optimization objectives, and measurement methodologies differ substantially (Kalkunte et al., 2024).

1. Terminological scope and principal meanings

The term is used in at least four recurring senses in the cited work.

Usage of “CBAND”	Meaning	Representative source
C-band optical networking	Baseline optical backbone spectrum pool, extended to C+L	(Kalkunte et al., 2024)
C-BASS / C-Band All-Sky Survey	5 GHz all-sky intensity and polarization survey	(Taylor, 2018)
Cellular band	Conventional cellular spectrum contrasted with TV white space	(Liu et al., 2011)
Channel bonding	Aggregating multiple channels into one wider channel	(Kai et al., 2017)

In optical networking, the C band is the constrained starting point of elastic optical backbone capacity, and the literature explicitly treats it as the baseline resource pool before migration to C+L-band operation (Kalkunte et al., 2024). In radio astronomy, by contrast, C-BASS is a survey instrument and data product at 5 GHz, designed for component separation of Galactic foregrounds from the CMB (Taylor, 2018). In cellular-spectrum studies, CBAND = cellular band refers to the conventional band used before opportunistic access to TV white space is introduced (Liu et al., 2011). In WLAN and CRSN research, “CB” denotes channel bonding, and the relevant question is not a frequency designation but the optimal selection of contiguous channels under contention or primary-user activity constraints (Kai et al., 2017).

This suggests that any technical use of “CBAND” must be interpreted locally from context rather than assumed to name a single technology class.

2. C band as an optical backbone resource pool

In optical transport, the cited work treats the C band as the original, limited-spectrum operating mode of the elastic optical backbone and studies how it can be extended to C+L-band operation. The core problem is dynamic traffic management under continuous traffic growth, with traffic requests carrying source, destination, required rate, minimum acceptable rate, traffic type, compression factor, maximum delay, and holding time. Each request is represented as

$$(s, d, \gamma, \overline{\gamma}, q, \phi, \delta, \tau).$$

Both the C-band-only and C+L bands scenarios are evaluated, with each band having 133 channels spaced by 37.5 GHz (Kalkunte et al., 2024).

The proposed mechanism is the Delay-Aware and Compression-Aware (DACA) provisioning algorithm. It first attempts provisioning in original form using standard RSA based on k-shortest path routing and First-Fit spectrum assignment. If no feasible slot exists, DACA exploits service flexibility: delayable traffic can be postponed until spectrum becomes available, and compressible traffic can be admitted at a lower data rate, thereby consuming fewer frequency slots. The paper states: “Compression lowers the required data rate $\gamma$ by a factor of $\phi$, thereby reducing the number of FSs required to provision the request. If compression does not help, the request is blocked.” The principal performance metric is Blocking Probability (BP), interpreted operationally as the fraction of traffic requests that cannot be provisioned (Kalkunte et al., 2024).

The quantitative outcome is framed as an increase in information-carrying capacity through reduced BP. In the C-band-only case, NDNC yields a blocking probability of 13.4% of total network traffic, while DACA reduces that by 36% relative to NDNC. When the L band is added, NDNC BP drops to 4.6%, and DACA improves this by 74% relative to NDNC. The same study reports that average utilization during peak hours in C band is around 50–60%, whereas in C+L with NDNC it drops to about 40%, and with DACA it drops further to about 30%. QoT is re-estimated dynamically using GSNR, with lightpaths re-evaluated every $t^p$ during peak hours and every $t^o$ during off-peak hours; the QoT estimators are ML-based and separate for C-band-only and C+L-band operation (Kalkunte et al., 2024).

A plausible implication is that, in this literature, “CBAND” in optical systems denotes not merely a wavelength interval but the baseline congestion domain against which traffic-flexibility mechanisms and multi-band upgrades are evaluated.

3. C-BASS: 5 GHz survey instrumentation and foreground science

A second major meaning is C-BASS, the C-Band All-Sky Survey, a ground-based, all-sky, full-polarization survey at 5 GHz designed to provide a low-frequency anchor for Galactic foreground characterization in support of WMAP, Planck, and future CMB B-mode polarization experiments (Jones et al., 2018). The survey maps Stokes $I, Q, U$ over the full sky, with 45 arcmin FWHM resolution and sensitivity to all angular scales from the instrument beam to the full sky. The northern telescope is at Owens Valley Radio Observatory (OVRO), California, and the southern telescope is at Klerefontein, South Africa; their optics are matched so that the two surveys can be combined into a consistent all-sky product (Jones et al., 2018).

The instrumental architecture combines a continuous-comparison radiometer for total intensity and a correlation polarimeter for linear polarization. In the polarization formalism used by the survey,

$$\langle |E_R|^2 + |E_L|^2 \rangle = I,\qquad \langle E_R E_L^* \rangle = (Q+iU)/2,$$

$$\langle E_L E_R^* \rangle = (Q-iU)/2,\qquad \langle |E_R|^2 - |E_L|^2 \rangle = V,$$

with

$E_{R,L} = (E_x \pm iE_y)/\sqrt{2}.$

The choice of 5 GHz is explicitly motivated by the fact that synchrotron remains bright while Faraday rotation is much smaller than at lower frequencies. One paper states that 5 GHz is simultaneously the highest frequency at which the foreground polarization will be clearly detected all across the sky, and the lowest frequency at which the confusing effects of Faraday rotation and depolarization can be robustly corrected (Jones et al., 2018).

The survey’s scientific role is primarily foreground separation. Simulations and early analyses show that including a 5 GHz C-BASS data point strongly constrains synchrotron spectral parameters, reduces degeneracies with free-free, AME, and CMB, and improves recovery of the CMB amplitude in both temperature and polarization (Jew et al., 2019). More specifically, in total intensity, when synchrotron curvature is fixed to zero, improvement factors for the synchrotron spectral index are typically about 1.8 to 6.4, free-free emission measure improves by roughly 1.4 to 5.1, and AME peak frequency constraints improve by up to about 3.6. In polarization, improvement factors for synchrotron amplitude are often tens to more than 100, and total error volume reductions range from roughly 50 to 2,000,000 depending on pixel and model. However, the same work states that additional low-frequency data in the 10–30 GHz range are needed to constrain synchrotron curvature properly (Jew et al., 2019).

C-BASS data are also used directly in astrophysical component-separation studies. In the first Galactic quadrant, two months of preliminary C-BASS North data at an effective intensity center frequency of 4.76 GHz, 43.8 arcmin FWHM resolution, and preliminary 5% calibration uncertainty were used to measure synchrotron spectral indices of $\beta = -2.65 \pm 0.05$ between 0.408 GHz and 5 GHz and $\gamma$0 between 1.420 GHz and 5 GHz in an off-plane region. After subtracting an RRL free-free template, the Galactic plane yielded $\gamma$1 between 0.408 GHz and 5 GHz, with $\gamma$2% free-free contribution at 5 GHz in the plane region studied. The same analysis reported diffuse AME detected at 4σ between 5 GHz and 22.8 GHz, and AME significances of 4.4σ for W43, 3.1σ for W44, and 2.5σ for W47 (Irfan et al., 2015).

Other C-BASS applications extend to external galaxies. For M31, C-BASS North total-intensity maps at 4.76 GHz combined with data from 408 MHz to 857 GHz yielded a flux density of about $\gamma$3 Jy, a preferred synchrotron spectral index of $\gamma$4, no strong evidence for curvature, and AME detected at $\gamma$5 with $\gamma$6 Jy peaking near 30 GHz (Harper et al., 2023). In the North Celestial Pole region, the 4.7 GHz map was limited by source confusion at about 0.6 mK rms, and replacing the Haslam 408 MHz synchrotron template with the C-BASS 4.7 GHz template changed the inferred AME amplitude by less than 10%, supporting the conclusion that a strong harder synchrotron component is not dominant above 5 GHz (Dickinson et al., 2018).

4. Cellular and CBRS uses of C-band and “CBAND”

In cellular-spectrum studies, CBAND is used explicitly to denote the conventional cellular band before TV white space is introduced. In the proportional-fair allocation problem of cognitive cellular networks, the system considers $\gamma$7 users $\gamma$8 and compares rate allocation in CBAND versus TV white space. The downlink CBAND baseline is

$\gamma$9

The optimal downlink result is that the TV-band set $\phi$0 must contain the highest-geometry users, whereas for the uplink $\phi$1 must contain the lowest-geometry users. The asymmetry is explained by the power disparity: in the downlink the paper assumes $\phi$2 dBm for TV-band fixed devices versus about $\phi$3 dBm in the cellular band, while in the uplink mobile transmit powers are both about 20–23 dBm (Liu et al., 2011).

A different but related literature uses C-band to refer to mid-band 5G-NR spectrum. In outdoor measurements in central Auckland, the compared systems were C-band: 3.60–3.70 GHz, center 3.65 GHz and mmWave: 26.65–27.45 GHz, center 27.05 GHz. The C-band deployment used 192 antenna elements, 64 RF chains, and zero-forcing (ZF) baseband processing, while the mmWave system used 320 antenna elements with analog beamforming (Shafi et al., 2020). The measured performance showed a maximum link distance of 1730 m for C-band versus 640 m for mmWave, about 10 dB higher SNR for C-band at the same CDF point, approximately 3% data loss in C-band versus 23% in mmWave, and a richer rank distribution with the maximum number of data streams to one UE equal to 8 for C-band and 4 for mmWave. Approximately 50% of the time, the C-band rank was greater than 2 (Shafi et al., 2020).

Within U.S. spectrum sharing, the CBRS band spans 3.55–3.7 GHz and sits adjacent to commercial C-band 5G beginning at 3.7 GHz. In a real deployment in South Bend, Indiana, the study documented both co-channel interference (CCI) among GAA users and adjacent channel interference (ACI) from C-band 5G. One measured edge-of-band case, PCI 194 on 3690 MHz, experienced about 16% throughput degradation when a nearby Verizon C-band UE was simultaneously transmitting; reassigning that PCI from 3690 MHz to 3560 MHz improved median throughput from 9 Mbps to 21 Mbps and median RSRQ from -14 dB to -11 dB (Tusha et al., 2024). In a separate coexistence analysis involving a ship-borne naval radar, the CBRS band (3550–3700 MHz) was studied under a radar INR protection criterion of $\phi$4 dB. The results suggested that 30 km protection distance ensures the required INR protection criterion with > 0.9 probability even in the co-channel case, while adjacent-channel operation can reduce the required protection distance to about 1 km (Krishnan et al., 2017).

These usages show that, in communications engineering, “CBAND” may denote either a legacy cellular spectrum pool, a practical 5G coverage layer, or a regulated shared band adjacent to higher-power C-band services.

5. Channel bonding as “CB” in WLANs and cognitive radio sensor networks

A further sense arises where CB stands for channel bonding, especially in IEEE 802.11ac WLANs and cognitive radio sensor networks (CRSNs). In the WLAN literature, channel bonding means aggregating multiple 20 MHz basic channels into 40 MHz, 80 MHz, or 160 MHz channels. The cited work distinguishes Static Channel Bonding (SCB), where transmission occurs on the full assigned bandwidth only if all channels are idle, from Dynamic Channel Bonding (DCB), where a WLAN contends on its primary channel and then uses the largest contiguous idle subset containing the primary channel (Kai et al., 2017).

The throughput analysis is built from a continuous-time Markov chain (CTMC). If WLAN $\phi$5 has mean backoff $\phi$6, attempt rate

$\phi$7

and mean packet transmission duration $\phi$8, then

$\phi$9

The stationary distribution is

$t^p$0

and the throughput of WLAN $t^p$1 is

$t^p$2

with $t^p$3 under the ideal-channel assumption (Kai et al., 2017).

The optimization result is that, for “all-inclusive” DCB networks, the maximal system throughput is obtained by channel allocations with the minimum overlap $t^p$4. In illustrative examples, the normalized throughputs satisfy

$t^p$5

The paper then formulates the allocation problem as an integer nonlinear program and solves it with a Branch-and-Bound Method (BBM). For $t^p$6, the legal widths are $t^p$7, $t^p$8, corresponding to 20, 40, 80, and 160 MHz. The fitted approximation

$t^p$9

uses $t^o$0 and $t^o$1. The claimed complexity is $t^o$2 when $t^o$3 and $t^o$4 when $t^o$5 (Kai et al., 2017).

In CRSNs, the relevant question is not contention among WLANs but protection of primary radio (PR) users. The proposed schemes are Remaining Idle Time aware Channel Bonding (RITCB) and RITCB-IP. PR activity is modeled as an ON/OFF continuous-time Markov renewal process, with exponential ON and OFF durations: $t^o$6 For each channel $t^o$7, the paper defines

$t^o$8

RITCB selects contiguous channels with the longest remaining idle time, while RITCB-IP adds a pre-transmission PR check and is reported to reduce harmful interference to zero in the evaluation. Both RITCB and RITCB-IP are described as having linear complexity $t^o$9, whereas PRACB is $I, Q, U$0 (Bukhari et al., 2017).

This literature uses “CBAND” only indirectly through “CB,” but it is a recurrent meaning in engineering discourse and is semantically distinct from the various senses of C band.

6. Satellite, propagation, and fixed-satellite-service uses of C band

Several cited papers use C-band in satellite and propagation studies. In integrated LEO satellite–terrestrial systems, a map-based experimental study considers operation at $I, Q, U$1 GHz, LEO altitude 550 km, and receiver height 1 m, with path loss modeled as

$I, Q, U$2

The free-space term is

$I, Q, U$3

Using a real 3D map of London, ray tracing in MATLAB, and model fitting, the study finds strong dependence of satellite-to-ground channel gain loss on elevation angle, building density, and average building height (Nguyen-Kha et al., 2024).

The fitted logarithmic models are given explicitly. For elevation angle,

$I, Q, U$4

with $I, Q, U$5 and $I, Q, U$6. For building density,

$I, Q, U$7

with $I, Q, U$8, $I, Q, U$9, and $$\langle |E_R|^2 + |E_L|^2 \rangle = I,\qquad \langle E_R E_L^* \rangle = (Q+iU)/2,$$0. For average building height,

$$\langle |E_R|^2 + |E_L|^2 \rangle = I,\qquad \langle E_R E_L^* \rangle = (Q+iU)/2,$$1

with $$\langle |E_R|^2 + |E_L|^2 \rangle = I,\qquad \langle E_R E_L^* \rangle = (Q+iU)/2,$$2, $$\langle |E_R|^2 + |E_L|^2 \rangle = I,\qquad \langle E_R E_L^* \rangle = (Q+iU)/2,$$3, and $$\langle |E_R|^2 + |E_L|^2 \rangle = I,\qquad \langle E_R E_L^* \rangle = (Q+iU)/2,$$4. In dense urban areas, non-line-of-sight points can fall to around $$\langle |E_R|^2 + |E_L|^2 \rangle = I,\qquad \langle E_R E_L^* \rangle = (Q+iU)/2,$$5 to $$\langle |E_R|^2 + |E_L|^2 \rangle = I,\qquad \langle E_R E_L^* \rangle = (Q+iU)/2,$$6 dB, and near-blocked cases can reach around $$\langle |E_R|^2 + |E_L|^2 \rangle = I,\qquad \langle E_R E_L^* \rangle = (Q+iU)/2,$$7 to $$\langle |E_R|^2 + |E_L|^2 \rangle = I,\qquad \langle E_R E_L^* \rangle = (Q+iU)/2,$$8 dB (Nguyen-Kha et al., 2024).

In fixed-satellite-service operation, an extended C-band VSAT system in Karnataka, India, uses 6.875–6.9465 GHz uplinks and 4.650–4.7215 GHz downlinks via Intelsat-3A. The paper models a single one-way satellite link with Rectangular 16-QAM, a 36 MHz transponder, 7.2 m hub antenna, 1.2 m VSATs, and hub HPA power of 350 W or 600 W TWTA. The free-space losses are given as 221 dB uplink and 217 dB downlink, and BER results include 0.1236 with phase error, 0.5001 with frequency error, and 0.00052 after phase/frequency compensation (Surekha et al., 2012).

These studies do not use “CBAND” as an acronym in the same way as C-BASS or channel bonding, but they reinforce the frequency-centric meaning of C band as an operating regime whose engineering challenges include blockage, coexistence, amplifier nonlinearity, and BER sensitivity.

7. Cross-domain themes, misconceptions, and technical significance

A common misconception is that CBAND refers to a single technology or a single frequency interval. The cited work does not support that simplification. In some papers it denotes an optical spectrum pool centered on backbone congestion management (Kalkunte et al., 2024); in others it names a 5 GHz astrophysical survey instrument (Jones et al., 2018); elsewhere it signifies the conventional cellular band before TV white-space offloading (Liu et al., 2011), or an instance of channel bonding as a MAC/PHY bandwidth aggregation strategy (Kai et al., 2017).

Another misconception is that all uses of C band imply similar propagation or interference behavior.