Skip to content
New
Latest News || 27 October 2026: Dr Soumava Mukherjee's invited talk at Queen's University Belfast joined device-level hardware innovation with channel-level control. || 22 October 2026: I received a Certificate of Appreciation for serving as a reviewer for IEEE IMAS 2026. || 15 October 2026: I presented a poster on beam-domain ML fingerprinting for RIS-aided mmWave localisation at the Centre for Wireless Innovation's annual symposium. || 15 October 2026: CWI's annual symposium brought research and industry perspectives on 6G, ISAC, optical wireless, AI, and future networks to Queen's University Belfast. || 14 October 2026: Erik Nordell of Ericsson visited our Keysight Lab for a live RIS-assisted video streaming demonstration.
Tarek Hassan
Knowledge Hubintegrated sensing and communication4. Principles, Waveform Design & Trade-offs

integrated sensing and communication learning note

4. Principles, Waveform Design & Trade-offs

The signal-processing principles behind ISAC, including waveform design, sensing metrics, communication metrics, and unavoidable trade-offs.

At a glance

  • Three ways to design the waveformRadar-centric, communication-centric (such as OFDM) or joint: each starts from a different priority.
  • Power has to be splitWhen the user and the target sit in different directions, every watt sent to one is missing at the other.
  • Two scorecardsRate, BER and latency for communication; resolution, detection and estimation error for sensing.

How ISAC Works

ISAC works by designing one radio system to produce useful communication observations and useful sensing observations.

At the receiver side, the signal may be processed in two ways:

  • A communication receiver estimates symbols, codewords, and data.
  • A sensing receiver estimates delay, Doppler, angle, and target strength.

Sometimes these receivers are physically the same device. Sometimes they share only part of the infrastructure.

References for this section3

Waveform Design Strategies

References for this section3

Radar-Centric ISAC

A radar waveform is modified to carry data. This protects sensing performance but may limit communication flexibility.

References for this section3

Communication-Centric ISAC

A communication waveform, such as OFDM, is reused for sensing (see OFDM Sensing earlier in this track). This is attractive because it fits cellular systems, but the waveform may not have ideal radar ambiguity properties.

References for this section3

Joint Waveform Design

The waveform is designed from the beginning for both tasks. This is more powerful but more complex.

  • High data rate: does not automatically mean accurate sensing.
  • Clean radar pulse: may be unsuitable for flexible multi-user data.
  • Design task: choose a communication/sensing operating point for the actual service target.
References for this section3

Sensing Metrics

Common sensing metrics include:

  • Range resolution.
  • Velocity resolution.
  • Angle resolution.
  • Detection probability.
  • False alarm probability.
  • Estimation error.
  • Radar signal-to-interference-plus-noise ratio.
References for this section3

Communication Metrics

Common communication metrics include:

  • Data rate.
  • Spectral efficiency.
  • Bit error rate.
  • Outage probability.
  • Latency.
  • Energy efficiency.
  • User fairness.
References for this section3

The Fundamental Trade-off

The best waveform for communication is not always the best waveform for sensing.

For communication, random-looking data-bearing signals can be good. For sensing, carefully shaped ambiguity functions and predictable probing signals are often useful. ISAC must balance these requirements.

References for this section3

Beamforming Trade-off

  • Communication beam: points toward the user.
  • Sensing beam: illuminates the target.
  • Different directions: split power, time, beams, or spatial degrees of freedom; then measure both outcomes.

Splitting one beam between a user and a target

1. Formula and symbols
w=ρ au+1−ρ as∥ρ au+1−ρ as∥w=\frac{\sqrt\rho\,a_u+\sqrt{1-\rho}\,a_s}{\|\sqrt\rho\,a_u+\sqrt{1-\rho}\,a_s\|}
gu=∣auHw∣2/N,R=log⁡2(1+γ0gu)g_u=|a_u^Hw|^2/N,\quad R=\log_2(1+\gamma_0g_u)
  • ρ\rho: beam-mixing setting
  • aua_u: Steering vector toward the user
  • asa_s: Steering vector toward the sensing target
  • ww: normalized shared weights
  • NN: elements
  • gug_u: user-direction gain
  • γ0\gamma _{0}: reference linear SNR

Target gain uses the same projection toward as.

beam pattern (dB)0-10-20-30USecho SNR vs user rateuser rate →↑ echo SNRU = user, S = sensing target
2. Work with the current inputs
w=ρ a(θu)+1−ρ a(θs)∥ρ a(θu)+1−ρ a(θs)∥,R=log⁡2 ⁣(1+γ0∣aH(θu)w∣2N),SNRs=SNR0 ∣aH(θs)w∣2N\mathbf{w} = \frac{\sqrt{\rho}\,\mathbf{a}(\theta_u) + \sqrt{1-\rho}\,\mathbf{a}(\theta_s)}{\left\lVert \sqrt{\rho}\,\mathbf{a}(\theta_u) + \sqrt{1-\rho}\,\mathbf{a}(\theta_s)\right\rVert}, \qquad R = \log_2\!\left(1 + \gamma_0\frac{|\mathbf{a}^{H}(\theta_u)\mathbf{w}|^{2}}{N}\right), \qquad \mathrm{SNR}_s = \mathrm{SNR}_0\,\frac{|\mathbf{a}^{H}(\theta_s)\mathbf{w}|^{2}}{N}
3. Result
User SNR17.9 dBrate 5.98 bit/s/Hz
Echo SNR11.3 dB
Directions apart55°the beam must split
A 16-element array must serve a user and illuminate a target in different directions with the same total power. The weights mix a beam toward each, w ∝ √ρ·a(θ_user) + √(1−ρ)·a(θ_target). Left: the resulting beam pattern. Right: every possible split traced as a curve (the Pareto frontier); the dot is your current choice. Moving along the curve, one side improves only by taking from the other. Reference SNRs with the full beam: 20 dB for the user, 15 dB for the echo.
References for this section3

Self-Interference

If a full-duplex ISAC transceiver transmits and receives simultaneously, the transmitted signal can leak into the sensing receiver and overwhelm weak echoes. Self-interference cancellation is therefore a major practical issue.

References for this section3

Practical ISAC Design Checklist

  • Choose waveform: radar-centric, communication-centric, or joint.
  • Define sensing and communication metrics.
  • Allocate power between users and targets.
  • Design beamforming and receive filters.
  • Handle clutter and self-interference.
  • Decide how sensing feedback improves communication.
  • Validate under mobility, blockage, and hardware impairments.
References for this section3

Takeaway

ISAC is not simply "radar plus communication." It is a joint resource-allocation and signal-processing problem where spectrum, power, beams, waveform structure, and hardware are shared between two objectives.

References for this section3

Check your understanding

0 of 4 answered

  1. Which design reuses a cellular waveform such as OFDM for sensing?

  2. In the trade-off lab, what happens when the user and target directions are far apart?

  3. Why is self-interference a major problem for a full-duplex ISAC transceiver?

  4. Which is a sensing metric rather than a communication metric?

Complete references and further reading