I'd like to share an elegant solution that avoids using htmlentities and doesn't require manually listing all character replacements. This is the php translation of this post.

function removeAccents($str) {
    return preg_replace('/[\x{0300}-\x{036f}]/u',"",normalizer_normalize($str,Normalizer::FORM_D));
}

$a = "joaoaaeeA";
$b = "joãoâàéèÀ";

var_dump(removeAccents($a) === removeAccents($b));

Output:

bool(true)

Just convert the accented characters to their unaccented counterparts and then compare the strings. The function in my answer will remove accents for you.

function removeAccents($string) {
    return strtolower(trim(preg_replace('~[^0-9a-z]+~i', '-', preg_replace('~&([a-z]{1,2})(acute|cedil|circ|grave|lig|orn|ring|slash|th|tilde|uml);~i', '', htmlentities($string, ENT_QUOTES, 'UTF-8'))), ' '));
}

$a = "joaoaaeeA";
$b = "joãoâàéèÀ";

var_dump(removeAccents($a) === removeAccents($b));

Output:

bool(true)

Demo