---
title: "Did commercial deployment of Lucky's adaptive equalizer raise voiceband data rates to 9600 bps (4× the ~2400 ceiling), and is that figure in any primary source?"
type: "question"
status: "open"
date_raised: "2026-07-09T00:00:00.000Z"
tags: ["adaptive-filtering","bell-labs","telecommunications","primary-source-verification","quant-verification"]
---


Raised while promoting the capture behind
[[claim-voiceband-distortion-barrier-capped-data-at-2400-bps]] and
[[claim-lucky-1965-adaptive-equalizer-steepest-descent-transversal-filter]]. The
capture carried a third claim, flagged `[unverified-quant — needs primary]`:
secondary sources (Wikipedia, ETHW-derived bios) report that commercial deployment
of the adaptive equalizer enabled a jump to **9600 bits per second — 4× the prior
~2400 bps ceiling**. That figure was **not** located in Lucky's 1965 paper and was
left in the inbox rather than promoted, because a specific "Nx" multiplier is
exactly the kind of quantitative claim the sourcing floor requires Tier 1–2 to
carry.

Two things to establish before this number is load-bearing:

1. **Is the figure real and correctly attributed?** 9600 bps voiceband modems
   (e.g. the CCITT V.29 family) arrive in the *1970s–80s*, well after Lucky's 1965
   paper — so "the equalizer enabled 9600 bps" may be compressing a decade of
   incremental modem engineering into a single attribution. Determine what rate
   Lucky's *own* 1965/1966/1967 equalizer work actually claimed or demonstrated.

2. **Where is the primary?** Candidates: Lucky's own follow-up BSTJ papers
   (1966/1967 general-purpose automatic equalizers), Lucky, Salz & Weldon's 1968
   textbook *Principles of Data Communication*, or the relevant CCITT/Bell modem
   specifications. Read directly and pin the rate to a dated primary before any
   note rests on it.

Medium priority — the mechanism and the 2400 bps barrier are already Tier-1
confirmed; this only concerns the downstream throughput number.