edVirtus Reference

Who is Erdal Arıkan?

Erdal Arıkan (1958-): The Information Theorist Who Discovered Channel Polarisation

Erdal Arıkan is a Turkish electrical engineer and information theorist who invented polar codes. His 2009 construction was the first explicit family of codes with efficient encoding and decoding proved to achieve the symmetric capacity of binary-input memoryless channels as block length grows.

Polar coding answered a foundational question left by Claude Shannon's channel coding theorem: can a capacity-achieving code be described and decoded with manageable complexity rather than shown to exist through a random ensemble? Arıkan's answer was to transform many equal noisy channels into synthetic channels that become sharply unequal.

Education and Bilkent University

Arıkan was born in Turkey in 1958. He studied electrical engineering at the California Institute of Technology and completed graduate work at MIT in the mid-1980s, developing expertise in information theory, communications, and sequential decoding.

After serving on the faculty at the University of Illinois, he joined Bilkent University in Ankara in 1987. There he pursued the theory and applications of error-correcting codes while teaching generations of communications engineers.

From Existence to Construction

Shannon proved that, for rates below channel capacity, codes exist whose error probability can be made arbitrarily small. The proof did not provide one universally simple recipe for constructing and decoding such codes at practical block lengths.

Richard Hamming supplied early constructive error correction, Robert Gallager developed low-density parity-check codes, and turbo codes brought iterative systems close to capacity. Arıkan sought a different combination: an explicit recursive transform, a rigorous capacity proof, and algorithms whose complexity grows nearly linearly with block length.

Channel Polarisation

The polar transform combines independent uses of a binary-input channel and then views the result as a sequence of synthetic bit-channels. Recursion makes some synthetic channels increasingly reliable and the rest increasingly unreliable, even though the total mutual information is conserved.

Information bits are placed in the reliable positions. The unreliable positions are frozen to agreed values known to encoder and decoder. As the block length increases, the fraction of useful positions approaches the symmetric capacity of the underlying channel.

Successive-Cancellation Decoding

A basic polar decoder estimates bits in sequence, using the received observations and earlier decisions to update the likelihood of the next bit. The recursive structure supports encoding and successive-cancellation decoding with complexity proportional to N log N for block length N.

That efficiency accompanies a risk: an early wrong decision can propagate through later estimates. The asymptotic proof is powerful, but ordinary systems use finite blocks, where performance depends on code construction, channel knowledge, and decoder design.

Making Short Polar Codes Competitive

Successive-cancellation list decoding keeps several candidate paths instead of committing immediately to one. A cyclic redundancy check can then help select among the surviving candidates. These developments greatly improved finite-length performance while increasing memory and computation.

Rate matching, punctured codes, shortened codes, hardware parallelism, and improved reliability ordering extend polar coding to varied payloads and channel conditions. The engineering code used in a standard is therefore a developed family, not simply the asymptotic algorithm copied from one paper.

A Precise Role in 5G

Polar codes were selected for important control-channel functions in 5G New Radio, where relatively short messages require strong reliability and flexible rates. This brought Arıkan's theoretical construction into a worldwide communications standard within a decade of publication.

It is misleading to say that polar codes carry all 5G traffic. Low-density parity-check codes protect the main user-data channels, while polar coding serves control information. The division reflects a trade-space analysis among block length, throughput, latency, decoder complexity, and implementation maturity.

Beyond Channel Coding

Polarisation has been extended from channel coding to source coding, multi-user problems, secrecy, and other information-theoretic settings. The recurring technique is to transform a collection of moderate instances into extremes that are easy to classify and use selectively.

The work also revealed connections with older Reed-Muller code structures and stimulated new mathematical questions about the speed of polarisation, scaling, universality, and optimal finite-block construction.

Recognition and Legacy

Arıkan received the IEEE Richard W. Hamming Medal, delivered the 2019 Claude E. Shannon Lecture, and in 2026 received the ACM Paris Kanellakis Theory and Practice Award for channel polarisation and polar codes. He remains a professor at Bilkent University.

His legacy joins proof with deployment. Polar codes do not make every other channel code obsolete, but they provide a rare explicit bridge from a capacity theorem to efficient algorithms and a global standard, while clarifying why asymptotic optimality and practical finite-length performance are distinct questions.