The phased array antenna is the single most transformative technology in radar history. By replacing mechanical rotation with electronic beam steering, phased arrays achieve beam agility measured in microseconds rather than seconds, enabling simultaneous search and track, adaptive waveform control, and electronic counter-countermeasure techniques impossible with mechanically scanned antennas. This article traces the evolution of phased array technology and examines the fundamental principles that govern its performance.

Beamforming Fundamentals

A phased array forms and steers its beam through controlled phase (and sometimes time delay) shifts applied to each element. For a uniform linear array of N elements spaced d apart, the far-field radiation pattern is the product of the element pattern and the array factor. The array factor for a beam steered to angle θ0 is the sum of contributions from each element with progressive phase shift θn = n × (2π/λ) × d × sin(θ0). This coherent summation produces a main lobe at θ0 with beamwidth approximately λ/(Nd cos θ0) radians.

The phase shifter is the core enabling component. Analog phase shifters using switched delay lines, vector modulators, or reflective phase shifters provide continuous or quantized phase control with insertion loss typically 3–8 dB. Digital phase shifters using discrete bits (typically 5–7 bits, providing 11.25° to 2.8° resolution) are simpler to control but introduce quantization lobes that degrade sidelobe performance. True time delay (TTD) units, which delay the signal rather than phase-shifting it, eliminate beam squint — the frequency-dependent pointing error that occurs when phase shifters are used with wideband waveforms.

Passive vs. Active Arrays

Passive electronically scanned arrays (PESA) use a single high-power transmitter and receiver, distributing the signal to and from the elements through a beamforming network. While simpler and cheaper than active arrays, PESAs suffer from beamforming network losses (typically 3–5 dB in both transmit and receive paths), single-point transmitter failure vulnerability, and limited waveform flexibility.

Active electronically scanned arrays (AESA) place a transmit/receive (T/R) module behind each element, containing power amplifier, LNA, phase shifter, and T/R switch. This distributed architecture eliminates the lossy beamforming network (improving sensitivity by 3–5 dB on both transmit and receive), provides graceful degradation as individual modules fail, and enables multiple simultaneous beams through digital beamforming. The cost, complexity, and thermal management challenges of AESA are offset by its overwhelming performance advantages, making it the standard for modern defense radar.

Grating Lobes and Element Spacing

When the element spacing exceeds λ/2, the array pattern develops grating lobes — additional main lobes at angles where the phase progression adds coherently. These parasitic beams waste transmit power, degrade receive sensitivity through noise integration, and create angle ambiguities. In practice, the maximum element spacing is d < λ/(1 + |sin θmax|), where θmax is the maximum scan angle. For ±60° scan, this limits spacing to approximately 0.53λ.

This spacing constraint drives the module miniaturization challenge: at X-band (10 GHz), λ/2 = 15 mm, and the entire T/R module, including connectors, cooling, and control electronics, must fit within a 15 mm square lattice. At millimeter-wave frequencies, the constraint becomes even more severe, requiring highly integrated module designs with wafer-scale packaging.

Digital Beamforming

Digital beamforming represents the apex of phased array evolution. In a fully digital array, each element has its own ADC and DAC, and all beamforming is performed in the digital domain. This eliminates analog beamforming networks entirely, provides maximum flexibility (arbitrary numbers of simultaneous beams with independent waveforms), and enables advanced processing such as space-time adaptive processing (STAP) and adaptive nulling at the element level. The challenge is data throughput: a 1,000-element array with 500 MHz bandwidth per element produces 2 TB/s of IQ data. Advances in ADC technology, FPGA processing, and high-speed interconnects are making fully digital arrays practical for an expanding range of applications.